Kru revisited, Kru revealed Lynell Marchese Zogbo ... - Llacan

Kru revisited, Kru revealed
Lynell Marchese Zogbo
Welcome to the KRU (“crew”) language family
• Very small under 12 million speakers
• Limited to Côte d’Ivoire and Liberia, with Kru
settlements Freetown Ghana
• Past documentation rare (Koelle 1800’s,
Delafosse 1904, Thomann 1906)
• Dense forest, no kings, no classes, no masks (at 1 pt)!
endangered? Krus learn other langs, not reverse
• Recent surge in research from late 60’s til present:
-published grammars: Thomann, Innes, G. Zogbo;
-doctoral dissertations: Kokora, Grah, Egner, G. Zogbo,
Saunders, Marchese, Thalmann;
-UQAM group: Kaye, Koopman, et al.;
-SIL research: Bentinck, Leidenfrost, etc.
-ILA theses and publications
-Kru Atlas 1979 and Tense/Aspect and the development of auxiliaries in Kru (1979)
• standard Niger-Congo features: noun class remnants, verbal extensions, a *CVCV word
structure, labio-velars and at least one implosive stop, ATR vowel harmony, nasalized
vowels in one half of family, and three to four level tones, as well as common lexical roots
CLASSIFICATION
Within Niger-Congo
• independent by Westermann and
Bryan
• under Kwa by Greenberg (refuted
by Welmers, Williamson)
• linked to Gur (Bennett and Sterk)
• Williamson and Blench (2000) have
Kru under Proto Volta-Congo
Needs a lot more work!
Compelling : noun class suffixes
Proto-Niger-Congo
________Kordofanian
Proto-Mande-Atlantic-Congo
Atlantic___________________________Mande
Proto-Ijo-Congo
____________Ijoid
Proto-Dogon-Congo
Dogon___________
Proto-Volta-Congo
*West Volta-Congo East Volta-Congo=Proto Benue Kwa
KRU Kwa…Benue…Bantoid
Pre Gur-Adamawa
Senufo

Internal classification
Delafosse (1904) proposed division into two groups, Bakwé and Bété, now called Western and Eastern,
divided by Sassandra river: There are several isolates:
• Kuwaa and Dewoin in Liberia (confirmed)
• Aizi in Côte d’Ivoire (confirmed)
• Sεmε in Burkina Faso (not confirmed)
West-East division demonstrated by correspondences, isoglosses & phonological feature:nasalized V’s
t : s PWK *tu PEK *su
‘tree’ tū Nyabwa sū Neyo
tū Wobé sū Godie
tū Gueré sū Dida Y
tūgbὲ Tepo sū Bete GU, Bete GB
tū Klao sū Kouya
tuū́ Bakwe
cu Bassa *tu > cu.
ɲ: ŋ PWK *ɲ PEK *ŋ
‘name’ ɲlɩ́ Nyabwa ŋlɩ́ Bete (D)
ɲɩńɩ́ Wobé ŋlɩ́ Dida (Y)
ɲnɩ́ Guere ŋnɩ́ Kouyo
ɲnέ Klao ŋlʉ H Godié
ɲrɩ́ Bakwe

‘woman’ PWK *ɲ PEK *ŋ
ɲnɔ́ Krahn (T) ŋlɔ́ Neyo
ɲrʋgba Tepo ŋwlɔ́ Godie
ɲnɔ́ Guere ŋlɔ́ Dida Y
ɲnáákpʋ̄ Wobe ŋɔ́ Dida L
ŋlɔ̄ Bete
ŋwnɔ́ Kouya
ŋlʋ́ʋ́ Bakwé
Note that here Bakwe shows a Western form for ‘name’ and an Eastern one for ‘woman’.
gb: gw
‘dog PWK *gbe M PEK *gwɩ
Nyabwa gbē Godie gwe-yi
Wobe gbē Dida go-yi
Guere gbē Bete GU gwɩ̄
Tepo gbì Bete GB gwē
Klao gbè Bete D gwɩ̄
Bassa gbē Kouya gwɩ̄
Bakwé and Neyo show fricatives for this lexical item. Neyo has vɩ or vɔ M and Bakwe, fɛ L.
Numerous isoglosses:
Western Eastern
‘fire’ PWK *nε PEK *kosu
Nyabwa nε̄ Neyo kòsù
Wobé nε̄ Godie kòsú
Guere nε̄ Dida (Y) kòsù
Tepo nā Bete-G kòsù
‘tooth’ PW *nynɩ H PKE *gle M
Nyabwa ɲnɩ́ Neyo grē
Wobé ɲnɩ́ /nynɩε Godie glē
Guéré ɲnɩ́ɛ̀ Bete (GU) glē-yì
Tepo ɲέε̄ Bete (GB) glε̄
Klao ɲέ Bete (D) glē-yī
Bassa ɲεńέ Kouya gla 1
Bakwé glὲ
Questions: Eastern Neyo? Dida? border languages Bakwé, Kouya ; Western subgroup Wè vs Kru ;
1 This is the plural of gle in many languages.

PROTO-PHONOLOGY
*CVCV-(C) V & probably *CVV ; current CLV and syncopated syllable types are derived
*Four level tones, with a few examples of 4 > 3, innovated modulated tones resulting from toneconsonant
interaction and consonant voicing; tone carries a high functional load in the lexicon and in
the grammar: distinguishing 1-2 nd person (sg and pl), perfective-imperfective, negation
egg
Proto Kru Initial consonants (Marchese unpublished, 1976) current analysis
p b c g kp p t k kp
b d ɟ k gb b d g gb
m n ɲ ng ɓ l w
s m n ? ?
z w s
kw, gw derived from CVV c, j, ny derived? g > ɟ gie> ɟ ie
l realized as flap l, r, n, ɗ ???relationship bt l-d-ɗ,
c, s > h in several Kru languages *y
k > f, s > z z, s > f v? g > gh group of Bete
*ɓ Ny Wob Guer Tep Bass Bakw Kouy BeD BeG Godi Ney De
Leg ɓʋ̄ ɓʋ̄ ɓʋ̄ ɓʋ̀ ɓo bɔ̄ʋ́ ɓʋ̄ ɓʋ2 ɓō ɓɔ̀ɔ́ ɓō
Two stages Proto Kru I (oldest ) and Proto Kru II
Proto-Kru II ‘bush’ kwálá or kwlá < Proto Kru I **kʋlá
‘moon’ *cʋ h < Proto Kru I **kí-ʋ (Kuwaa: kewu)
‘dew’ *ɟ lù < Proto Kru I **dilù
‘breast *ɲiti < Proto Kru I ?
n…………………………..ng(w) ………………… …w
‘mouth’ nё, ne Godie, Neyo ngwε Kouya, wɔn Klao, Bassa
ngo Bete D wɩ TepoKrumen
nguo Wobe, Guere wɔɩ Dewoin
‘hear’ nú Neyo, Godie, Dida ngwɔ Grebo wɔn Tepo, Klao, Bassa
nʋ̄ Bete, Kouya
CompareWilliamson and Bench (2000:41)
*Western Sudan Kordonfanian *Proto-Kru
#nu, ‘hear’ -eenu ‘ear’ *nú? ‘hear’

#nu, -nua ‘mouth’ *uungu *nV? ‘mouth’
VOWELS
Proto Kru oral vowels PWK Nasalized
ɩ ʋ ɩn ʋn
e o
ε ɔ εn ɔn
a an
*vowel harmony everywhere: retracted but some +height
*predictable nasalization after nasal C or N dropping
*divide: nasalized Vs in West, almost none in East; all nasalized V’s except e & o
Western (nasalized vowels)
Nyabwa Wobe Guere Klao Bassa
Tail gʋn M gʋn M gʋn M wɔn M vɔn M
Arm sʋn M sʋn M sʋn M sɔn M
Two sɔn H sɔn H sɔn H sɔn H sɔn H
Eastern (oral vowels)
Neyo Godie Dida BeteGGB Koouya
tail gʋ go M go M go-yì
Arm sɔ sɔ sɔ sɔ
Two sɔ H sɔ H mɔsɔ BH sɔ sɔ H
Complementary distribution in Western Kru
Klao (Singler): voiced stops and oral Vs /dan/ [nan] MH ‘drink’ vs /da/ ‘call’MH, never
[dan], never [ma], but distinction with voiceless stops: pi ‘cook’ pin M ‘load (on head)’
Did Proto-Kru have nasalized V’s or not? If so, what triggered de-nasalization in Eastern Kru?
Positing proto nasalized vowels and no nasal consonants (Le Saout, Bole-Richard, Vydrine for
Mande) doesn’t work for East Kru. Could Kru have borrowed nasalized vowels from Mande??
Central vowels : innovated in Eastern Kru (Bete, Godie) but not in Dida and Neyo:
i ï U
ɩ ʉ ʋ
e ə O
ɛ ʌ ɔ
a
Bakwé has central vowels, is it genetic or areal? Why this innovation? SMande=central V’s
Note weird fact: where central vowels occur, there are no nasalized Vs!!!

PROTO-MORPHO-SYNTAX
TAM
• Currently SVO, but much OV typology:
-ɔ lï sʉkʌ (SVO) ɔ yi sʉkʌ lï (S AUX O V)
‘he eats rice’ ‘he will eat rice’
-GEN N with or without associative*a (alienable/inalienable)Nyabwa gbe á kpá bone of dog
-N Postposition
-OV noun compounding
• Kru is 100% suffixing: plurals in nouns & verbs (extensions: applic, passive, valency)
• No serial verbs but conjoined propositions with AUX sequential from ‘come’
• Comparison: expressed by *proto ‘pass’ construction: *X is good pass Y
• Topic-comment style with front shifted focus, affirmative focus particles within clause
Aspect dominates: perfective and imperfective reconstructed for Proto Kru with perfective
basically an unmarked verb stem (but there is a pervasive low tone throughout the family).
Proto imperfective is the marked form, reconstructing to *NP a V-e (Marchese, 1982)
Which is older? Both persist. In some languages a has suffixed onto NP, producing an
imperfective pronoun set. In others a is lost. In many languages –e reduces to mid tone. Did
low perfective tone marking innovate to provide a clear distinction (dissimilation)? Did a proto
tense marker get re-analyzed, then reduce to low tone (contra expectation tense > aspect)?
There is a proto periphrastic progressive * he is-at bathe-place ‘he is bathing’.
Tense is secondary. Some evidence of recent and remote past tense suffixes: #a or #i ‘recent’
(Godie, Dewoin) and #o/wo, wʌ (remote). But extensive tense innovation derived from temporal
adverbs most extensively in the Kru cluster (Klao, Cedepo, TepoKroumen, Bassa, etc). (See
Marchese, 1984). The innovative tenses range from 1 to 5, highest in Grebo (Innes, 1969):
(i) né dú blà. ʻI pounded rice ʼ.
I pound rice
(ii) né dú-da blà. ʻ I pounded rice the day before yesterday ʼ.
(iii) né dú-d́ blà. ʻI pounded rice yesterday. ʼ
(iv) né dú- blà. ʻ I pounded rice today.ʼ
(v) né dú-a blà. ʻ I will pound rice tomorrow.ʼ
(vi) né dú-d blà ʻI will pound rice the day after tomorrow. ʼ
Auxiliaries express future, potential, perfect, sequential, negation, relatively recent development
and currently on-going : SV (O)V-nom S AUX (O) V, (1986)

‘come’ future, sequential, potential; ‘have’ future, conditional, ‘bring’? .perfective,
‘stop’, let go’ NEG markers. True reanalysis : Aux: no perf-imperf, take obj & tense suffixes
NOUN CLASSES and AGREEMENT
Proto Kru words were made up of *STEM + Class suffix
Vestiges seen in:
Pronouns:
• Pronoun systems 3 rd person
• Singular + plural forms : regular and irregular
• In some languages, in definite markers
• Remnants of concord
Primary distinction is human vs non human:
3 rd singular with hierarchy human > big animals > everything else
Human Non-human
Kuwaa (isolate) ɔ ɛ̄
Niaboua (Wè) ɔ̄ ɛ̄
Wobé (Wè) ɔ̄ ɛ̄
Grebo (Kru cluster, W) ɔ ɛ
Djabo ɔ ɛ
Kouya (East) ɔ we
Tepo Kru (Kru cluster, W) ɔ ɛ, o, ɔ
Neyo (E) ɔ ɛ, a, ʋ
Godié (E) ɔ̄ ɛ, a, ʋ
Bété (Daloa) (E) ɔ́ ɛ, a, ʋ 2
Dida (Yocoboué)
ɔ ɛ, a, ʋ
On the basis of these pronouns, we posit for 3 rd singular human *ɔ and three non-human classes
*ɛ, *a, *ʋ. In many cases there are clear semantic categories which link to class markers and or
remnants in other families. Currently, however, this system is described phonologically by all
(Grah, Kaye, Marchese, Sauder, Saunders, Thalmann, Werle, Zogbo): in Eastern Kru, NHum
words ending in front V’s take ɛ, central Vs, a, back Vs, ʋ, but final V reflects old noun class
marker.
In Godié nouns ending in front vowels belong to the *ɛ big animal class (many taking –a in the
plural):
mlɛ “animal” Ɉɛ “antilope
lʋɛ “elephant” glɛ “monky” (SP)

kaɓɛ “monkey”(SP) tlɛ “snake”
gbalɛ “hippopotamus” bɔlɛ “monkey”
ɓlɛ “buffalo” gwɛ “chimpanzee”
ɓaɓlɛ “sheep” ɓlɛ “antelope”
kpəkɛ “crocodile” Ɉɩ “panther”
dʋdʋzʋɛ “anteater” ɓlɩ “cow”
Compare other families for the word ‘goat’ 3:
Yoruba e-wúrɛ́
Efik e-bot
Igbo e-wu
Proto-bantu ɩN-boli
Kordofanian e-bonyi
Kru (Godié) wuli-ɛ (definite form)
Godié nouns ending in back vowels take the *ʋ pronoun. Note liquids and mass.
liquids non –solid masses
ɲʋ́ “water” jlʋ̀ “fog”
ɓlʋ “milk” vʋvɔlʋ “wind”
dlʋ “blood” gbaylʋ “smoke”
nʋ “alcoholic drink” ɓàɓʋ̀ “dust”
zo “soup” nyɔmʋ “air”
bubʋ “sweat” zùzʋ “spirit, shadow”
natural elements miscellaneous (these take ɩ in plural)
kòsu “fire” wlʋ́ “head”
lagɔ “sky, God” lʋ “song”
dʋdʋ “earth” blɔ “road”
glʋ “soil” ylʋ “day”
ylʋ “sun”
cʋ “moon, month”
Compare PB 3/4 mʋ/mi , PBenue-Congo ʋ/(ɩ)I, Togo Remnant *o/*i , the *o –prefix in Yoruba:
omi “water” ɔna “road” ɔbɛ “soup” orin “ song”
ɔrun “heaven” ošu “moon” ori “head”
The a class is not so definable, but includes food, instruments, small insects, etc. Examples from Godie :
sʉkʌ ‘rice’, nimlə ‘bird’, nyidə ‘cooking pot’; Bete (G): ɟá ‘wound’, mla ‘nose’, zɩa, ‘bird’.
Reconstruction of noun stems is difficult because the noun class V suffix has ‘taken its spot’:

*ní +ʋ ɲú ‘water’ *lʋ-ε ‘elephant’ *yVlV- ʋ ‘sun’ nV + ʋ ‘alcoholic drink’
Class concord or agreement is reflected across the family first and foremost in pronouns:
subject, object, possessive, relative, interrogative, without overt reference in discourse.
Within the NP, concord marks various elements, depending on the language:
noun +adjective + demonstrative + number (definite floating)
Godie ɓìtì kʌdɩ HL nɩ sɔ ‘those two big houses’
Nyukpɔ kʌdɔ nɔ nii mlε kʌdε. Nykpɔɔ -ɔ nii mlε nʌ, ɔ kʋ ɔ ɓutu kádʋ
‘man big DEM saw animal big Man-def REL saw animal SUB, he is-at his house big
The big man saw a large animal’. . ‘The man who saw an animal is in his big house.
Noun + Adjective + Demonstrative Noun + definite
nyʉ̄kpɔ̄ kʌ́dɔ̄ nɔ̄ ‘this big (great) man’ nyʉ̄kpɔ̄ + ɔ LM ‘the man’
ɓùtu kʌ̀dʋ nʋ ‘this big house’ li + ε lie ‘the spear’
mlɛ̄ kʌ̍d̍ɛ̄ nɛ̄ ‘this big animal’ ɓùtù + ʋ ɓùtùu ‘the house’
nmlə kʌ̍dʌ̄ nʌ̄ ‘this big bird’
ɓı ̀tı̄ kʌ̍dɩ̄ nɩ̄ ‘these big houses’
nyʉ̄kpà kʌ́dʋa nʋa ‘these big (great) men’
The same agreement system exists in Bakwé, as seen in the following
Noun + adjective ‘only’ 4
nyɔɔ ˈdoolɔ le seul homme
tatu ˈdoolʋ la seule porte
sapɛ ˈdeelɛ le seul poulet
ˈkpata ˈdöölä la seule chaise
For adjectives, the situation varies greatly from language to language. Some agree with class and
number as in Godie and Bakwe forms above, while some only agree in number. Where there is
adjective class agreement, only a subset of adjectives agree, often the words ‘big’, ‘white’, ‘red’ and
‘new’. As the class agreement diminishes, plural agreement can remain. But note that for Tchien
Krahn, Sauder reports the qualified nouns take the non-human plural marker5. True everywhere
nyɔ klàbá ‘an important person/ man’ (big) nyʋ klàbɩ ‘important people’
dε zʋ n ‘bad thing’ dɩ zʋɩn ‘bad things’ nyʋ zʋɩn ‘bad people’
Tepo Krumen6: ya ká yɩ kɩ ‘old pot/pots’
hru pεtú hri pεti ‘short road/roads’
yu yrayrʋ yuó yrayrɩ ‘new child/children
4
Examples from Csaba Leidenfrost.
5
Sauder, p. 46
6
Thalman. P. 251

The less the number of pronoun (class) distinctions, the less concord within the NP. The closer
the element to the head noun, the more likelihood of agreement (no 1 st , dem/def 2 nd ). The loss
of agreement in adj works its way through the lexicon, class marked colors & plurals resisting.
Language Dem Adj1 Def Adj2 Num
Godié X x/10 X x/4? 0
Bété (Guibéroua) X X X X/? 0
Neyo X X X X/? 0
Wobe X X/10 NR x/1 0
Kouya 0 X NR ? 0
Grebo X S ? 0 0
Kuwaa SG/PL X/1 X 0
KLao 0 X 0 0 0
Dewoin 0 ? 0 0 0
X = agreement present
0 = no agreement
Adj 1 agree in number
Adj 2 agree in class & number
Definite seems innovative
• The Kru system shows important links to human, large animal, and liquid-mass classes
in other NC languages.
• It ‘respects’ their “membership” even including what appears to be unusual members!
• It suggests that proto NC might have had less than more (noun classes)!
• The Kru system shows there are hierarchies of semantic categories in ‘reduction
schemes’ with certain categories persisting: human over everything else, animate (?)
over inanimate, big over little? Color over ?, Plural over class?
• How and why do class and concord systems develop and why do they fade? Do any other
NC language families show class agreement moving to phonological agreement?
Marchese, L. (1979, 1983), Kru Atlas, Abidjan: ILA , 1979, 1983.
______________ (1982) 'Basic Aspectual Categories in Proto-Kru' in JWAL, Vol. 12, No. 1. pp. 3-23.
______ (1984) 'Exbraciation in the Kru language family', Historical Syntax, lacek Fisiak (ed.). Mouton, pp. 249-270
______ (1984), 'Tense Innovation in the Kru language family', Studies in African Linguistics.
Vol. 15 No. 2, UCLA.
______ (1986) Tense/Aspect and the Development of Auxiliaries in the Kru Language Family, SIL-Universitiy of
Texas Press, Arlington, 1986, 301 pages. (UCLA Phd thesis under Welmers and Thompson, 1979)
-------(1988) 'Noun Classes and agreement systems in Kru: A Historical Approach', in Agreement in natural
language: Approach, theories and descriptions. Barlow and Ferguson (eds.), Stanford University Press.
______ (1989) 'Kru', in The Niger-Congo Languages. 1. Bendor Samuel (ed.), Lanham: University Press of
America, Inc. - SIL, pp. 119-13
Thanks to Csaba Leidenfrost, Philip Saunders, Peter Thalmann, R. Zogbo, J. Blewoue & especially Hortense Tebili!